1. Field of the Invention
The present invention relates to a superlattice infrared photodetector, which can be fabricated easily with molecular beam epitaxy, has the features of low power consumption, and small dark current. Furthermore, the working temperature to operate the detector under the background limited performance can be achieved by cooling down to the liquid nitrogen temperature. Moreover, the ratio of the photocurrent to the dark current can be improved effectively so that the working temperature for background limited performance increases greatly to even higher than 77 K.
2. Description of Related Art
The conventional quantum well infrared detector (QWIP) is designed for far infrared radiation detection (6-14 mm) which can be used at night vision, medical diagnosis or national defense missile systems, etc. The word xe2x80x9cquantumxe2x80x9d indicates the wave characteristics of the electrons. If the motion of the electrons is confined in space, the energy of the electrons will be quantized, and the resulting difference between the quantized energy levels falls in the infrared spectral region.
EP 275-150-A discloses a detector with an energy band structure illustrated in FIG. 1(a), which uses an energy band structure established by different material. In general, the quantum well and barrier regions are made of GaAs and AlGaAs respectively. In the quantum well, the electron is confined in space and two bound states are formed there. The electron in the lower energy level can transit to a higher energy level by the absorption of infrared radiation. The photo-excited electron has equal possibilities to move rightwards or leftwards so as to tunnel to an adjacent quantum well. To cause these electrons to move in the same direction, an external electric field must be added so that the electrons have a larger tunneling probability in the direction reverse to the electric field. The photocurrent due to tunneling effect of the photo-excited electrons will be detected for the electric field larger than a critical value for sufficient photo-excited electrons to be detected and therefore, the applied bias voltage is large.
In order to increase photocurrent, and avoid the aforesaid critical voltage, B. F. Levine et. al, in the paper with a title of xe2x80x9cHigh Detectivity D*=1010 cm/{square root over (Hz)}W GaAs/AlGaAs Multiquantum Well xcex=8.3 xc3x85 Infrared Detectorxe2x80x9d in App. Phys. Left. 53(4), Jul. 25, 1988, reduced the energy barrier so that the second energy level becomes a continuous one, as illustrated in FIG. 1(b) and thus, photoelectrons can be easily accelerated by electric field to arrive at a contact. In contrary, this will cause the dark current to increase dramatically. To reduce dark current, the barrier width is increased to a value of 50 nm, while the result is not preferred. The reasons are involved as follows. The dark current is induced from the physical mechanism of thermionic emission (in small bias), or thermally assisted tunneling (in large bias), and is primary determined by the difference between the Fermi level of the electron doping concentration in the quantum well and the energy height of the barrier. Both of the two physical mechanisms depend slightly on the width of the barrier. In particular, the thermal assisted tunneling is much related with the magnitude of the electric field instead of the applied voltage.
For the two quantum well infrared detectors described above, due to the restriction of the critical voltage and the increment of the barrier width, the bias range is within several volts for a detector with approximate 30 to 50 periods. Therefore, the power consumption must be taken into account if the heating effect due to the power consumption must be limited. Besides, if the dark current is too large in the operation point, not only the problem of the power consumption becomes serious, but also it is possible to affect the read-out circuit because the capacitance of the read-out circuit may not be large enough to withstand a large dark current. On the other hand, the most simple and cost-saving way for cooling these detectors is to use liquid nitrogen. Namely, it is preferably that the working temperature is about 77 K. In this situation, the cooling time of liquid nitrogen can be elongated due to the low power consumption of our detector.
In order to avoid the redundant power consumption, in general, it is expected that the detector should work in the background limited performance (BLIP). Namely, the photocurrent due to the background photon illumination is larger than the dark current. Under the background limited performance, the quantum well infrared detector must be cooled to a temperature at least lower than 65 K. Therefore, it is not suitable to cool the detector simply by liquid nitrogen (77 K). The dark current must be further reduced to increase the working temperature for background limited performance. The U.S. Pat. No. 5,198,682 discloses a way in which the doping concentration in adjacent quantum wells or the aluminum component in the energy barrier (i.e. the energy height of the barrier) is changed gradually so as to reduce the dark current. This way is executable. However, because of the multiple thick energy barriers, a strong electric field is necessary for the photocurrent to pass through these energy barriers, and this results in considerable power consumption.
In order to reduce the inappropriate power consumption in a quantum well infrared detector, the inventor of the present invention discloses a superlattice infrared photodetector (SLIP) in the Electrochemical Society Proceedings Vol. 99-22 (page 485-495, 1999), wherein the superlattice structure is illustrated in FIG. 1(c) with a thinner energy barrier, about several nanometers. The single wavelength detection achieved by the transition between electron states in multiple quantum well is extended to a multiple-wavelength detection (indicated by the arrow in FIG. 1(c)) by the transition between minibands formed by the overlap of the electron wave functions in the adjacent potential wells. In addition, because of the thinner energy barrier, the electrons in the second energy band can freely move rightwards or leftwards nearly without external bias, and therefore, the power loss can be reduced significantly. Besides, to reduce the dark current induced by the electrons in the first miniband, and to collect the photoelectrons in the second miniband, a blocking layer with a width of about 50 nm is added to the rear side of the superlattice, as shown in FIG. 1(d). It should be noted that the height of the energy barrier of the blocking layer must be higher than the bottom of the second miniband. U.S. Pat. No. 5,077,593 discloses a similar idea, but the height of the energy barrier of the blocking layer is required to be lower than the bottom of the second miniband. Our design has the advantages of reducing dark current greatly and selectively choosing the required photoelectrons. Namely, the wavelengths to be detected are tunable. This is a great different to the prior art. However, if the detector is to work in the background limited performance, from experience of the inventor, it should to be cooled below 60 K. It is because that the difference of the height of the energy barrier of the blocking layer with the Fermi level of the electron doping concentration in the superlattice is not sufficient large.
Another superlattice detector worth of mentioning is claimed by U.S. Pat. No. 5,352,294 as illustrated in FIG. 1(d). In that structure, an intermediate contact is added between the superlattice structure and the blocking layer so that the dark current flows out from this contact. However, for the photoelectrons to pass through the intermediate contact, the width of the intermediate contact must be thin enough, and the energy barrier height of the blocking layer must be small enough. Although this design can reduce the dark current, the manufacturing process becomes more complex. The dual contact structure is converted into a three contact one. Furthermore, the electrical connection to the middle contact is very difficult because of the small contact width.
Accordingly, the primary object of the present invention is to provide a superlattice infrared photodetector, which can be fabricated easily, has lower power consumption and smaller dark current. Furthermore, the working temperature to operate the detector under background limited performance can be achieved by cooling through liquid nitrogen. In addition, the superlattice structure are sandwiched between the front and rear blocking layers with sufficient height and width. The thickness is about 50 nm and the height of the energy barrier is 10 meV higher than the bottom of the second miniband of the superlattice structure. Thereby, as a photocurrent is generated, the dark current will reduce automatically. Therefore, the ratio of the photocurrent to the dark current can be improved effectively so that the working temperature for the background limited performance is increased greatly to even higher than 77 K.
The various objects and. advantages of the present invention will be more readily understood from the following detailed description when read in conjunction with the appended drawing.